ColinOldnCranky wrote:
"Crime against statistics"? Why is it wrong to report this? As you almost indicate they are not saying cause and effect, they are simply alluding to a correlation so you have appently read their conclusions. It is YOU who have then gone beyond what is in evidence in this modest study to append your own opinion, unsupported at this time by any explanation or data. Give them a break Zero, this is a GOOD thing.
I've got nothing against the
authors of the study, beyond the amusing theorising in the conclusion. The predictive power of their model I'm fine with, as that would fit many possible causes and does not appear to be an unreasonable model of the data they are using.
The
reporter who wrote the age article however reported "CYCLISTS riding without bike lights are three times more likely to be seriously injured in a crash than those who are lit up, a major study of cycling crashes in Melbourne has found", which is an absurd statement. The study is limited to people that are injured in crashes to the extent of requiring hospitilisation, and does not cover any of the population of riders in Melbourne, or most importantly the population of riders in Melbourne who crashed, or the population of riders with lights. Even the majority of crashes in the dataset are single vehicle accidents in daylight without even another vehicle involved - which cannot be explained by any rational understanding of the bicycle light.
It would be wrong to NOT report such data if it is available and can be done easily. It all adds to what is available to other researchers. (And to opinionated bicyclists and unicyclists.
)
